Discrimination vs. Generation: The Machine Learning Dichotomy for Dopaminergic Hit Discovery

TL;DR

This study compares machine learning and deep learning methods for discovering dopamine D2 receptor agonists, showing that generative models yield higher hit rates and led to identifying a potent novel compound.

Contribution

It introduces a comparative analysis of predictive and generative ML strategies for drug discovery, highlighting the effectiveness of generative models in identifying novel agonists.

Findings

Generative ML model achieved higher hit rate.

Discovery of Compound 1, a nanomolar dopamine D2 receptor agonist.

The pipeline effectively differentiates active from inactive ligands.

Abstract

Virtual screening plays a pivotal role in early drug discovery, traditionally dominated by physics-based methods. While these approaches offer detailed insights, they are often hindered by high computational costs, limited sampling, and forcefield inaccuracies. Advances in Machine Learning (Ml)and Deep Learning (DL) present resource-efficient alternatives, with approaches like predictive geometric ML (EQUIBIND) and generative geometric ML (DIFFDOCK)showing promise in enhancing both efficiency and predictive capability. Here, we compare these two strategies, retrospectively and prospectively, for identifying novel agonists targeting the dopamine D2 receptor. To complement DIFFDOCK's dual functionality in protein-ligand conformer generation and confidence estimation, we adopted a complementary atom-type-based confidence model for EQUIBIND. This pipeline, termed the discriminative model, integrates a featurization step and an XGBoost classifier to differentiate between active and inactive ligands. The top-ranked compounds from both models were evaluated using an ultrafast dopaminergic biosensor assay, dLight. Our results demonstrate that the generative model achieved a higher hit rate, notably leading to the discovery of Compound 1, a nanomolar dopamine D2 receptor agonist with a novel scaffold.